1. Introduction

This is an automatic report for RGB & MONO image fusion result. And this report is generated on Fri May 06 13:35:10 2016.

This paper show RGB & MONO image fusion experiment. This step is a component of supernight pipeline.

2. Algorithm

The input rgb image is the output of the rgb denoise and mono image is the output of the alignment. We take the following steps to do RGB & MONO image fusion.

3. Experiment

3.1. Fusion with un-denoised rgb image

The following are input rgb image without denoise and mono image.

and after our fusion algorithm, we get fusion image.

It is observed that, without rgb image denoise, the fusion image have many color noise. This may have side-effect for our supernight feature.

3.2. Fusion with denoised rgb image

So, we do another experiment. The input rgb image is a denoised image.

and after our fusion algorithm, we get fusion image.

3.3. Two result comparison

Let’s compare the different fusion result.

For a closer look, we crop a small piece from the image and enlarge it.

Through Comparative Analysis, we found the fusion image with rgb denoise didn’t have color noise anymore, its y channel noise come from mono image. And we will denoise the fusion y channel noise later.